Figure 4 - uploaded by Lokesh Chouhan
Content may be subject to copyright.
Plot between Actual and Predicted Trend of LSTM The above graph figure 3 is plot over the data having batch size 512 and 90 epochs. The prediction is shown by red line and the actual trend is shown by blue. The proximity of these two lines tells, how efficient the LSTM based model is. The prediction approximates real trend when a considerable amount of time has passed. The more thesystem is trained the greater the accuracy which will be attained IV. CONCLUSION Two techniques have been utilized in this paper: LSTM and Regression, on the Yahoo finance dataset. Both the techniques have shown an improvement in the accuracy of predictions, thereby yielding positive results. Use of recently introduced machine learning techniques in the prediction of stocks have yielded promising results and thereby marked the use of them in profitable exchange schemes. It has led to the conclusion that it is possible to predict stock market with more accuracy and efficiency using machine learning techniques.
Source publication
In Stock Market Prediction, the aim is to predict the future value of the financial stocks of a company. The recent trend in stock market prediction technologies is the use of machine learning which makes predictions based on the values of current stock market indices by training on their previous values. Machine learning itself employs different m...
Similar publications
Stock market prediction has always been a difficult process, most of the prediction rely solely on the data of the corresponding stock market. Relationship of gold and oil price with stock market performance has been proven significant in some major world stock index. Prediction of stock market price index using machine learning methods is expected...
We use machine learning methods to forecast individual stock returns and create long-short portfolios in the Brazilian stock market, using a rich data set including technical and fundamental predictive signals. We further develop an algorithm that combines long-short portfolios obtained with various machine learning methods such that (i) the risk c...
Citations
... In the realm of stock trading, ML has gained increasing prominence, with investment firms utilizing it for a large number of purposes, including predicting market fluctuations, analyzing customer behavior, and scrutinizing stock price dynamics [15]. Notably, one of the most critical applications of ML is in the domain of stock price analysis [16]. Nevertheless, predicting stock prices using ML is a formidable task, given the countless physical, psychological, and other dynamic factors that influence stock values. ...
Predicting stock market prices accurately is a major task for investors and traders seeking to optimize their decision-making processes. This research focuses on the comparative analysis of advanced machine learning (ML) techniques, particularly, the Long Short-Term Memory (LSTM) model and Autoregressive Integrated Moving Average (ARIMA) model for predicting stock market prices. The study enforces thorough data collection and preprocessing to ensure the quality and reliability of the historical stock price data, forming a robust foundation for the predictive models. The core contribution of this paper lies in its systematic and comparative analysis of these two models. A range of performance metrics, including Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), are employed to assess and contrast the predictive accuracy and efficiency of the LSTM and ARIMA models. The research findings indicate that the ARIMA model, contrary to expectations, outperforms the LSTM model in this study, achieving lower RMSE and MAE values. Specifically, the ARIMA model demonstrates a Test RMSE of 4.336 and a Test MAE of 3.45926, indicating its superior predictive accuracy compared to the LSTM model. Furthermore, the study sets its findings against the backdrop of existing literature by comparing the performance of its models with those reported in previous research. This comparison shows better results achieved by our stock market prediction models. By addressing limitations observed in prior studies and demonstrating practical applicability, this research contributes to advancing stock market prediction methodologies, offering valuable insights for investors and traders.
... There is no doubt that the economy of any country is strongly affected by the performance of its stock market [1]. During recent years, the financial markets around the world have faced significant fluctuations and uncertainties, so the existing uncertainty about the return on investments has worried many financial analysts and investors [2]. As investors point out, uncertainty is considered as the most important factor in determining the actual price of financial assets [3,4]. ...
... This means that the dataset obtained needs to be accurate, large, and detailed. Netflix is one of the largest entertainments in the U.S. Its reputation provided a large non-financial variable that the research wanted to cover so that the prediction method that this paper use could potentially applied to other company with high social influences [5,6]. Therefore, this paper will be researching Netflix's stock price trend. ...
As big data and Artificial Intelligence (AI) are more and more getting into society, people are beginning to ask for predictions in all fields to make prevention. With the technological improvements, the stock market has become more and more predictable as compared to before. The finance field, especially the stock market field, was being put on for making predictions. This paper evaluates the pros and cons, accuracy, and precision when making predictions using the ARIMA model and the Linear Regression model on Netflix stock price trends. The result shows that both models perform moderately. From that, this paper also found that the Linear Regression Model seems to be performing better in the short term and the ARIMA model predicted that it might perform better in the term of long term. This paper proved that these models could play a role in the field of finance but are still restricted to the dataset’s size and the application of the model. Hopefully, this paper will find well to finance related members, give and inspire some advice to them.
... Machine learning, employing a base of technologies, is the most recent method in stock market prediction. It involves basing the current stock market index on the training of past values (Parmar et al., 2018). ...
This paper refers to a previous study of Barua (2020), which shows the impact of the pandemic on aggregate demand and supply. The novelty and the purpose of this study is to examine the fundamental impact arising from the uncertainties created by the COVID-19 pandemic caused by the Indonesian government restrictions policy (spectrum) on stock returns of the sample companies listed on the Indonesia Stock Exchange (IDX) with a minimum of 200 trading days in 2020 from 2016 to 2020. This study used multiple regression analysis and cross-sectional for the cumulative abnormal return (CAR). The result shows that the fundamentals tested had no significant effect on cumulative abnormal returns. On the other hand, only the current ratio showed a significant effect on Spectrum 1 and 4; Spectrum 1 had a positive effect while Spectrum 4 had a negative effect, and the debt service coverage ratio (DSCR) had a slightly significant effect on Spectrum 1. The study results indicate that each spectrum had different levels of uncertainty, which gave rise to different perspectives for each investor. This study provides a perspective for investors to see different levels of uncertainty on the current ratio (CR) and DSCR as determinants of fundamentals for investment consideration.
... An analysis of the use of several ML algorithms useful in forecasting the future values of equities in the financial sectors has been performed (Obthong et al., 2020). For the purpose of forecasting stock price fluctuations, Parmar et al. (2018) constructed two models: regression and LSTM. According to the results, LSTM outperformed regression in terms of prediction accuracy. ...
Prediction of the stock market is a challenging and time-consuming process. In recent times, various research analysts and organizations have used different tools and techniques to analyze and predict stock price movements. During the early days, investors mainly depend on technical indicators and fundamental parameters for short-term and long-term predictions, whereas nowadays many researchers started adopting artificial intelligence-based methodologies to predict stock price movements. In this article, an exhaustive literature study has been carried out to understand multiple techniques employed for prediction in the field of the financial market. As part of this study, more than hundreds of research articles focused on global indices and stock prices were collected and analyzed from multiple sources. Further, this study helps the researchers and investors to make a collective decision and choose the appropriate model for better profit and investment based on local and global market conditions.
... They discover that, when compared to other algorithms, the 3-fold model performs best based on two error measures [2]. Presenting Ullah's Prediction on Google Stock Prices They forecasted the stock price for the first two months of 2017 using deep learning, using the Recurrent Neural Network (RNN) as a deep learning model, and they were able to do so with an accuracy of 87.32% [3,4]. For the purpose of classifying texts, Shah compares the performance of the Logistic Regression, Random Forest, and KNN models. ...
The prediction of stock prices is an intriguing and difficult study issue. Due to its frequent offers of quick rewards and low risk of loss, the stock market is currently regarded as a prestigious trading sector. Data mining and business researchers find the stock market to be an ideal environment due to its extensive and dynamic information sources. In this paper, the data of google stock price from 2018-1-2 to 2023-3-31 is imported. As our classification algorithms, the authors independently selected and contrasted super vector regression, random forest, and K-nearest neighbor. Following test, analysis, and comparison of these classifiers and calculate the mean square error of test data, the final result was reached. According to the result, the random forest has the smallest MSE value, so it is the best prediction and uses this method to predict the stock price of next ten days. After using random forest to do the stock price prediction, we find that the close price has the increasing trend for next ten days, which is good for the investors.
... To avoid losses and gain maximum pro its, investors search for techniques and tools to predict stock market movements (Ali et al., 2021;Ali et al., 2023;Sahu et al., 2023). However, predicting the stock market is tricky because of its dynamic, stochastic, and uncertain nature (Khan et al., 2020;Parmar et al., 2018). Two primary analytical methodologies are employed in forecasting stock market trends: technical analysis and fundamental analysis. ...
Predicting the movement of the stock price index has remained a challenging task for financial analysts. They are complex, given how the nation's economic, social, and political conditions affect stock market predictions. The present study is designed to build an efficient model using machine learning and taking monthly data from February 2004 to December 2020. Long- and short-term memory and backpropagation methods of an artificial neural network are utilized to predict the Karachi Stock Exchange (KSE) movement by using twenty-six economic, social, political, and administrative indicators. The model developed in this study to predict the movement of the KSE 100 index gained 99 percent accuracy. The predictions' results showed that the KSE 100 index would remain stagnant at around 40,000 points till September 2023. We compared actual and predicted values from January 2021 to July 2022 to validate our developed model. The results showed that the model developed in this study could be used to forecast stock market trends. The research predicted the KSE 100 index with 99% accuracy using LSTM and backpropagation in a neural network. The KSE 100 should stay around 40,000 points until September 2023. Despite their promise, machine learning forecasts in financial markets require continuing research, market dynamics evaluation, and liability disclosure. A thorough validation procedure and the most recent KSE 100 index prediction using LSTM and backpropagation algorithms demonstrate how machine learning may improve stock market forecasting. The study's accurate projections and consistent data through September 2023 are crucial for investors and financial experts in the Karachi Stock Exchange's changing environment.
... As features for the ML models, the work used the Relative Strength Index, Stochastic Oscillator, Williams Percentage Range, Moving Average Convergence Divergence, Price Rate of Change, and On Balance Volume. Parmar et al. [9] used the Yahoo Finance dataset to predict stock prices using Open, High, Low, Close, and Volume information. 20% of the dataset was used for testing, while the remaining 80% was used to train models. ...
There are a number of possible advantages in utilizing trading indicators and machine learning to predict the direction of stock prices. It is crucial to remember that stock price prediction is difficult by nature and that there is no way to ensure accuracy. The financial markets are extremely information-rich, dynamic, and complex. Large datasets may be processed and analysed by machine learning algorithms far more quickly than by people, which makes it possible to spot patterns or trends that might not be immediately obvious. By using historical price and volume data, the algorithms can be trained to identify patterns that could predict future moves. Because they offer insights into possible market moves, predictive models can help with risk management. Because ML models are always learning from fresh data, they can adjust to changing market conditions. This flexibility is essential in markets where a variety of factors impact the market. Several machine learning models have been used in this experimental effort to monitor the direction of stock prices, and the outcomes are extremely encouraging.
... To improve stock price predictions, machine learning was applied in the analysis. A special LSTM model outperformed traditional strategies, showing that more accurate methods are conceivable [13]. The flowchart's methodical approach ensures a thorough evaluation procedure, enabling us to find insightful information that advances our comprehension of algorithmic performance as shown in Figure 1. ...
... Least-square support vector machine (LS-SVM) was used to predict daily stock prices, and the Particle Swarm Optimization (PSO) algorithm was used to optimize (LS-SVM) [5]. Regression and Long short-term memory (LSTM) networks were used to predict the stock prices, opening, closing, maximum, minimum prices, and trading volume utilized [6]. Machine learning classifiers were applied to perform prediction in the stock market, using stock data, social networks, and news, where the random forest obtained the best result [7]. ...
The financial market generates a large volume of data daily, allowing the increasing use of machine learning algorithms in building predictive models for the stock market. In this environment, time is a crucial factor since stock prices change daily, so the training time of models is a critical factor. This paper proposes a method to optimize the overall training time of 5 reinforcement learning algorithms that predict the weights of each stock in a stock portfolio. Experiments were conducted by varying the number of algorithms executed simultaneously. In addition, the computational characteristics of each algorithm were analyzed concerning the use of memory and processing. From the proposed combination of running the algorithms concurrently, it was possible to reduce the total training time by 33% compared to running the algorithms sequentially. Moreover, this execution led to a commendable 15% decrease in energy consumption.
